Introduction
Stop runs are often described as a story about technical levels and trader psychology. That is only half the picture. The other half is structural: liquidity is not evenly distributed through time, and it is not isolated within a single market. When volatility rises abruptly, the first positions to fail are usually those that depend on stable correlations, continuous liquidity, and tight funding conditions. What exposes weak hands quickly is not merely a price level being touched; it is a volatility shock propagating through intermarket linkages, compressing risk capacity and forcing synchronized de-risking.
This matters for institutional and academic audiences because the mechanism is repeatable. It is less about predicting the next episode and more about building an account architecture that remains coherent when the cross-market plumbing tightens. Process precedes profit; structure carries outcome.
Core Idea
A stop run is best understood as a liquidity event with informational ambiguity. In normal conditions, many participants manage risk with rules that assume tradable liquidity near recent prices. When price moves quickly, those assumptions break. Stops, margin constraints, and risk limits convert discretionary intent into non-discretionary order flow. The market then experiences a short-lived imbalance: forced selling begets lower prices, or forced buying begets higher prices, until sufficient liquidity is induced.
Intermarket relationships amplify this. A volatility spike in one venue changes hedging demand in another, which changes funding needs elsewhere. Brunnermeier and Pedersen formalise this as the interaction between market liquidity and funding liquidity: when funding conditions tighten, dealers and leveraged participants reduce balance sheet usage, which reduces market-making capacity, which further worsens price impact. In practice, this is how a move that begins as a localized repricing becomes a cross-asset shock.
Market microstructure adds a second layer. Liquidity is state-dependent. Kyle shows how price impact emerges from the interaction of order flow and the market’s ability to absorb it. In stressed conditions, the same order size produces larger impact because the depth behind the best quotes is thinner and replenishes more slowly. When volatility rises, some liquidity providers widen spreads or step back, not because they “fear levels,” but because inventory risk and adverse selection risk increase.
The behavioural layer is not optional. Kahneman explains how loss aversion and narrow framing can cause traders to capitulate at precisely the wrong time, especially when P&L is evaluated too frequently. Yet even disciplined participants can be forced out if their risk architecture is procyclical: if leverage, position sizing, or hedges are calibrated to tranquil correlations and stable vol.
Market Reflection
Intermarket relationships are often taught as a diversification benefit. In reality, they are also a transmission channel. Correlations are not constants; they are conditional on regime, liquidity, and policy. BIS notes that stress episodes can be characterised by sudden repricing of risk and reduced market liquidity, with cross-market spillovers that are difficult to hedge in real time.
Several linkages are especially important during stop runs and volatility shocks.
First, volatility markets and underlying markets interact mechanically. When implied volatility rises, option dealers’ hedging flows can become more convex, increasing the sensitivity of hedging demand to price changes. This can create feedback loops in which the underlying moves, hedges adjust, and the adjustment itself adds to the move. While the details depend on positioning, the general point is robust: volatility is not merely a forecast; it is a traded state variable that can change the marginal buyer and seller.
Second, rates and funding conditions matter even for assets that appear unrelated. When short-term funding costs rise or collateral haircuts increase, leveraged strategies reduce exposure. The IMF documents how, in acute stress, even traditionally liquid markets can experience severe dislocations as investors scramble for cash and balance sheet capacity becomes scarce.
Third, cross-asset risk models can synchronize behaviour. Value-at-Risk and volatility targeting are widely used because they are operationally simple and auditable. But they can also be procyclical: higher volatility reduces risk budgets, which forces selling, which can further raise volatility. The Bank for International Settlements has repeatedly highlighted how risk management practices can amplify cycles when many institutions respond similarly to the same signals. BIS provides a useful framing of these dynamics in the context of market functioning.
Finally, the efficient market perspective is not contradicted by these episodes. Fama argues that prices reflect information, but the path of adjustment can still be discontinuous when liquidity is scarce and constraints bind. In other words, stop runs can be consistent with informational efficiency while being operationally violent.
Practical Discipline
The practical question is not how to avoid volatility shocks; it is how to avoid being forced into the worst possible execution when they occur. That requires “order before increase”: a portfolio should earn the right to scale by demonstrating that it can survive adverse states without improvisation.
Begin with a simple premise: if your strategy relies on stable correlations, you must stress those correlations. If your strategy relies on continuous liquidity, you must assume gaps. If your strategy relies on tight spreads, you must assume spread widening. These are not pessimistic assumptions; they are governance assumptions.
Decision hygiene matters because stress compresses time. Under pressure, teams revert to defaults. If defaults are ambiguous, the account becomes a behavioural experiment at the worst moment. Kahneman emphasises that humans are poor at consistent probabilistic reasoning under stress; robust processes exist to reduce the degrees of freedom when emotions and time pressure rise.
Risk should be framed as exposure to states, not just instruments. Intermarket linkages mean that a portfolio can be “diversified” by asset labels while being concentrated in one risk state: short volatility, long liquidity, long carry, or long funding stability. Brunnermeier and Pedersen provide the conceptual bridge between these states and the mechanics of forced deleveraging.
Account-Level Translation
Theory becomes useful only when it becomes enforceable. Intermarket linkages and stop-run dynamics translate into three account-level disciplines.
The account rule is a pre-committed constraint on scaling. Exposure may increase only after the account demonstrates stable behaviour across a defined set of stress proxies: higher realised volatility, wider bid–ask spreads, and correlation shifts across key markets relevant to the strategy. The rule is enforced operationally through a gating mechanism: if volatility or liquidity conditions exceed predefined thresholds, the account cannot add gross exposure, even if the view is strong. This is not a forecast; it is a structural check that prevents “doubling down” when the environment is most likely to punish fragile positioning.
The risk control is capital protection via convexity awareness and liquidity budgeting. The account explicitly budgets for liquidation cost and gap risk by limiting the portion of risk that depends on immediate execution. Practically, this means sizing positions so that a plausible adverse move combined with stressed spreads does not breach the maximum drawdown tolerance or trigger forced selling. The control is validated through scenario analysis that assumes cross-asset correlation convergence and funding stress, consistent with the mechanisms described by Brunnermeier and Pedersen and the crisis evidence summarised by the IMF . Capital is protected not by optimism about exits, but by ensuring the account can choose its exits.
The process discipline is a repeatable cadence that survives stress. It has three features: a fixed review schedule that does not accelerate with market noise; a single source of truth for exposures and liquidity metrics; and a pre-defined escalation path for when limits are approached. Under stress, the process does not ask for creativity; it asks for compliance. This is how “order before increase” becomes real: the account behaves the same way when headlines are calm and when volatility shocks propagate through intermarket channels.
Conclusion
Stop runs and volatility shocks are not mysterious ambushes; they are the visible surface of intermarket plumbing under strain. Weak hands are exposed quickly because they are positioned for a world of stable correlations, abundant liquidity, and forgiving funding. When those conditions change, risk becomes non-linear and exits become crowded.
A research-minded approach treats these episodes as design tests. The goal is not to predict the next shock, but to build an account that does not require perfect timing, heroic discretion, or continuous liquidity to remain solvent and coherent. If process precedes profit, then the most durable edge is not a clever view; it is a structure that can hold the view without being forced to abandon it at the worst possible moment.
References
Bank for International Settlements 2023, BIS Annual Economic Report 2023, BIS, Basel.
Brunnermeier, MK & Pedersen, LH 2009, ‘Market liquidity and funding liquidity’, Review of Financial Studies, vol. 22, no. 6, pp. 2201–2238.
Fama, EF 1970, ‘Efficient capital markets: A review of theory and empirical work’, Journal of Finance, vol. 25, no. 2, pp. 383–417.
International Monetary Fund 2020, Global Financial Stability Report: Markets in the Time of COVID-19, IMF, Washington, DC.
Kahneman, D 2011, Thinking, Fast and Slow, Farrar, Straus and Giroux, New York.
Kyle, AS 1985, ‘Continuous auctions and insider trading’, Econometrica, vol. 53, no. 6, pp. 1315–1335.
